The last common ancestor of Eugnathostomes


What kind of skeleton did the ancestral eugnathostome have? Did it have lots of dermal armor, little, or none?

We have four groups of well-known living and fossil gnathostomes to derive inferences from:

Placoderm grade vertebrates:

These have plenty of dermal bone, but in a unique pattern that makes it impossible (?) to compare with that of Osteichthyes.

Early stem chondrichthyans:

As we survey farther down the chondrichthyan stem, we find creatures that share chondrichthyan synapomorphies but increasingly share ancestral characters with other gnathostome groups. A sampling:

Ozarcus mapesae from Pradel et al. 2014.
Ozarcus mapesae (Carboniferous) A small symmoriid Pradel et al. 2014 described from CT scans. Although the braincase and amphistylic jaw suspension are typically chondrichthyan, the branchial arches are more like those of an osteichthyan. (Link to image from Pradel et al. 2014 modified for PaleoStories.)

Doliodus problematicus pectoral fin from Miller et al. 2003.
Doliodus problematicus (Early Devonian) Roughly 1 m long, it shows classic chondrichthyan synapomorphies. Interesting because its pectoral fins have spines of like those of the dorsal fins. Who else has spines like this?.

Gladbachus adentatus (Devonian) Little information in the literature, however in his 2013 podcast, Coates describes the results of his CT scans of Gladbachus, noting:

A general pattern emerges in which, the farther down the chondrichthyan branch we push:

Climatius from


(Late Ordovician - Permian)

Remember these guys:

Monophyly: Traditionally assumed. A potential synapomorphy: But this assumption invoked great stratigraphic incongruity. Some acanthodians seemed to exaggerate their typical features while others were similar to early Osteichthyes. E.G.:

Climatius from Carroll, 1988
Climatius: (Early Devonian)

Acanthodes from
Acanthodes: (Permian)

Difficulty: Generally speaking, one would expect that in an evolving monophyletic group, the more derived taxa would occur later in time than the more ancestral ones. This is called stratigraphic congruence. Acanthodians display the opposite pattern. The most freakishly spiny ones are also early.

Acanthodes neurocranium mandibular, and hyoid arches
Adding color to the mystery was the absence of preserved internal skull elements in any acanthodian but Acanthodes, whose skull seemed similar to that of osteichthyans. At the beginning of this century, the professional consensus was that Acanthodii and Osteichthyes formed a larger clade - Teleostomi.

Acanthodian surprise: Brazeau, 2009 described a second acanthodian braincase - Ptomacanthus anglicus. This creature's neurocranium retains many plesiomorphies reminiscent of placoderms and basal chondrichthyans. Brazeau's phylogenetic analysis revealed for the first time that: Moreover, we saw the general picture that the ancestral eugnathostome had:

And the people read it and rejoiced that the enigma had been resolved. But only until 2013.

Entelognathus primordialis by Brian Choo from Scale bar = 1 cm.

The Source of All Disquiet

Then came Zhu et al. 2013, reporting on Entelognathus primordialis, a placoderm-like creature with extensive bony cranial and thoracic armor and:

Now, with the addition of a heavily armored creature close to the ancestry of Eugnathostomata, the polarity was reversed. What if eugnathostomes arose from creatures with extensive armor? The lack of bony dermal plates in Chondrichthyes and Acanthodii might be a synapomorphy! Zhu et al.'s analysis recovered a phylogeny in which all acanthodians were stem chondrichthyans.

Adding color to this pattern:

If this is correct, maybe we should start reconsidering the long-discounted notion that other plates of the placoderm armor are homologous with osteichthyan dermal elements, as well. One dermal element is definitely present in both arthrodirans and osteichthyans - the parasphenoid - the dermal element flooring the sphenoid region of the neurocranium on the palate.